In order to study a claims database, it is handy to look at a worst-case scenario, in terms of size. Today, we’ll look at the U.S. Office of Personnel Management’s (OPM) Health Claims Data Warehouse (HCDW), a database of personal information generated from employees’ healthcare claims, as specified by the Affordable Care Act.
The data warehouse is a combination of both federal employees Federal Employee Health Benefit Program (FEHBP) and Obamacare subscribers (Pre-Existing Condition Insurance Service Plan, or PCIP), and is scheduled to maintain records for seven years. Since this was first proposed in 2010 and it has not been 7 years since inception, there has been no opportunity yet to explore how a purge would work or how frequently it would occur.
The purpose of this database? According to OPM’s director John Berry: “The warehouse will allow OPM to better understand the health of federal employees, as well as the cost and quality of care they receive. The warehouse will give OPM the ability to manage the program so that employees and tax-payers get the best value.”
Well, it looks like Mylan’s EpiPen price hike (from $57 in 2007 to $600 in 2016) may be the first real test of the government stepping in, but it seems to be from public outcry instead of business intelligence analysis spewing a report of the biggest price increases. Valeant’s cardiac drug Isuprel and Turing’s HIV drug Daraprim caused any federal legislation yet either (just inquiry) despite even larger price increases. The U.S. FDA has long banned imported drugs (which would immediately bring price relief) despite the fact that almost all vaccines have historically come from overseas.
Personal recorded information (as mentioned in 2010) for HCDW was stated to include patient’s name, Social Security number, date of birth and gender; contact information including phone number and address; provider charges and reimbursement amounts; and specific medical details including diagnoses and procedures. This sounds very much like an article I wrote about MIDAS, which dealt with security concerns about personal data in databases connected to the Affordable Care Act. While this data therefore contains personally identifiable information (PII) and personal health information (PHI), the door has been left open for possible research use once the data has been scrubbed into a deidentified format (as stated from the address in the next paragraph).
On September 12, 2015, upon a metadata update, was described on https://catalog.data.gov as follows: “The Health Claims Data Warehouse (HCDW) will receive and analyze health claims data to support management and administrative purposes. The Federal Employee Health Benefits Program (FEHBP) is a $40 billion program covering approximately 8 million eligible participants using more than 100 health insurance carriers. The HCDW will incorporate extensive analytical capabilities to support cost analysis, administration, design, and quality improvement of healthcare services provided to eligible participants.”
So the question remains: has the business intelligence built into the data warehouse in fact been guided to the originally–conceived purpose of understanding people’s health, the cost and quality of care they receive, and the ability to manage the program so that insureds leverage the best value? Until drug prices are brought under control, perhaps not. The same applies to procedure or equipment rates. The intelligence is there for potential analysis; the question is: is it being used?
The government website merely states that the metadata was created on August 30, 2014 and updated on September 12, 2015. At that rate, we may be ready for the next update any day now. Perhaps a spread across 2 years instead of 1 will be enough for rate changes or waste or rate variance to start showing up in the reports, and for business intelligence determinations to start to be made, to help optimize expenditures to reduce overall care expenses. We can only hope that data warehouse testing of report information will help solve the nation’s runaway costs, leading to a more managed solution.
Scott Andery is an expert marketer and author who specialize in software testing tools and resources.